Search Results (180)

Primary prevention is essential to reduce disease burden before clinical onset, yet it remains less systematically integrated into care than diagnosis and treatment. Although artificial intelligence (AI) is increasingly used in medicine, most applications have focused on secondary and tertiary prevention, including diagnosis, prognostic stratification, and disease management, while its role in primary prevention remains less defined. This narrative review examines current AI applications across four modifiable lifestyle domains relevant to prevention and healthspan promotion: nutrition, physical activity, sleep, and mental health. We synthesize evidence on machine-learning models, wearable-derived algorithms, computer-vision tools, just-in-time adaptive interventions, and conversational agents used in consumer, community, and hybrid clinical–digital settings. AI applications support postprandial glycemic prediction, automated dietary assessment, meal-planning adherence, sedentary-pattern detection, personalized exercise recommendations, adaptive behavioral nudges, sleep monitoring, circadian-aware recommendations, psychoeducation, stress-management support, and early identification of psychological vulnerability. Collectively, these tools may extend prevention beyond episodic clinical encounters toward continuous, personalized, and context-aware support. However, evidence remains limited by short follow-up, reliance on surrogate or engagement outcomes, digitally literate populations, and insufficient validation in real-world preventive-care pathways. AI is therefore a promising enabling technology for proactive, healthspan-oriented medicine, provided future studies demonstrate long-term effectiveness, equity, safety, and responsible implementation. Full article

Liquid biopsy has emerged as a transformative approach in modern oncology, offering minimally invasive access to tumor-derived biomarkers through the analysis of circulating tumor DNA, cell-free DNA, and extracellular vesicles such as exosomes. Unlike conventional tissue biopsies, liquid biopsy enables real-time monitoring of tumor dynamics, molecular heterogeneity, treatment response, and the development of therapeutic resistance. Recent advances in ultra-sensitive molecular technologies, including digital droplet polymerase chain reaction, next-generation sequencing, methylation profiling, and fragmentomic analysis, have substantially improved the sensitivity and specificity of circulating nucleic acid detection, facilitating their integration into precision cancer medicine. ctDNA analysis has demonstrated significant clinical utility across multiple malignancies, including lung, breast, colorectal, pancreatic, and prostate cancers, particularly in the identification of actionable genomic alterations, minimal residual disease, and mechanisms of acquired resistance. In parallel, cell-free DNA provides broader insights into tumor biology and systemic genomic alterations, while exosomes contribute additional layers of molecular information through the transport of nucleic acids, proteins, and signaling molecules involved in intercellular communication and tumor microenvironment modulation. The integration of artificial intelligence and machine learning approaches further enhances the interpretative power of liquid biopsy-derived datasets and supports the development of personalized therapeutic strategies. Despite these advances, important challenges remain, including low tumor fraction in early-stage disease, biological and technical variability, clonal hematopoiesis-associated false positives, assay standardization, and cost-effectiveness considerations. Nevertheless, the expanding clinical applicability of liquid biopsy technologies positions them as essential components of contemporary precision oncology. This review summarizes the biological foundations, analytical methodologies, current clinical applications, technological innovations, and future perspectives of circulating tumor DNA, cell-free DNA, and exosome-based liquid biopsies in cancer diagnosis, monitoring, and personalized treatment strategies. Full article

Exosomes are 30–150 nm extracellular vesicles that convey molecular information reflecting the physiological and pathological states of their source cells. In precision oncology, they function as a non-invasive “liquid biopsy,” enabling real-time monitoring of tumor dynamics and metastasis. However, extreme biofluid heterogeneity poses significant challenges for their isolation and analysis using conventional statistical approaches. This review aims to examine how artificial intelligence (AI), specifically machine learning and deep learning, transforms complex exosomal “noise” into actionable clinical insights. AI enhances exosome isolation, enables disease-specific biomarker identification, and predicts therapeutic responses with high precision. Integrating multi-omics data and single-exosome analysis enables AI-driven models to facilitate early cancer detection and therapeutic resistance monitoring. Despite challenges related to standardization and data privacy, the convergence of AI and exosome biology is poised to transform reactive cancer treatments into a proactive, personalized medical ecosystem. This approach also provides a framework for managing other complex systemic diseases. Full article

The survival probability of persons requiring rescue after large-scale earthquakes or landslides decreases rapidly with time, and nighttime search operations are constrained by limited visibility. Satellite remote sensing enables wide-area observation at night, but its limited spatial resolution restricts the identification of individual persons. This study develops an automated method for detecting persons in thermal infrared imagery captured by Unmanned Aerial Vehicles (UAVs) using the deep learning model Grounding DINO. To address the limited availability of Unmanned Aerial Vehicle (UAV)-based training data, thermal infrared imagery captured by fixed-point ground cameras was used for training. Spatial resizing and padding were applied to emulate UAV viewpoints and mitigate the ground-aerial domain gap. The proposed model outperformed the baseline on real-world datasets, with substantial Recall improvements in environments with low thermal contrast. An ablation study confirmed that spatial resizing, padding, and position-based augmentation each contribute progressively to detection performance. A comparison with the lightweight YOLOv8n and DETR-based RT-DETR detectors indicated that the Transformer architecture alone does not account for reliable detection in scenes with complex thermal noise. This robustness instead derives from the language-grounded semantic priors of Grounding DINO. An extended evaluation on nighttime imagery indicated that the proposed approach generalizes to nighttime acquisition. Full article

Gestational diabetes (GDM) is a frequent health problem associated with both short- and long-term adverse outcomes for mother and child. Standard management includes lifestyle interventions and, when necessary, pharmacologic therapy. However, the effectiveness and timely initiation of pharmacological therapy depend on accurate glucose monitoring. Continuous glucose monitoring (CGM) systems have emerged as valuable tools in diabetes care, providing real-time information on glycemic variability and enabling more individualized therapeutic interventions. In this narrative review, we explore the role of CGM in the early detection of dysglycemia, its diagnostic and prognostic value, and its ability to identify specific glycemic patterns during pregnancies complicated by GDM. We also assess its role in optimizing lifestyle interventions and guiding pharmacotherapeutic strategies. Current evidence suggests that CGM supports clinical decision-making and patient engaging by providing real-time glucose data. This facilitates earlier identification of hyperglycemic patterns, more precise treatment changes and improved glucose control. Furthermore, CGM use has been associated with improved neonatal and maternal outcomes. Despite these promising findings, barriers such as cost and limited access persist. Although the existing evidence remains relatively limited, it supports the integration of CGM into routine care of women with GDM as part of a comprehensive and personalized treatment strategy. Larger clinical trials are needed to fully understand the benefits and optimal use of CGM in GDM, as well as its impact on pregnancy outcomes, glycemic control and psychological well-being. Full article

Cardiogenic shock (CS) is the most lethal complication of acute myocardial infarction (AMI), with a 30-day mortality of approximately 40–50% despite early revascularization. Temporary mechanical circulatory support (tMCS) devices, including the intra-aortic balloon pump (IABP), microaxial flow pumps (MAFP) and veno-arterial extracorporeal membrane oxygenation (VA-ECMO), are used as adjunctive therapy in refractory shock, but evidence of a survival benefit is limited and often conflicting. The IABP-SHOCK II trial found no 30-day mortality reduction with IABP, supporting a Class III (no benefit) recommendation, whereas the DanGer Shock trial reported a 12.7% absolute mortality reduction at 180 days with the MAFP Impella CP in highly selected patients. In contrast, the ECLS-SHOCK and ECMO-CS trials showed no improvement in survival with early VA-ECMO and noted high complication rates. Real-world data reveal significant disparities between trial populations and clinical practice, highlighting limitations of current evidence, since many AMI-CS patients are older, in more advanced shock or have multiple comorbidities and would not meet typical randomized controlled trial (RCT) inclusion criteria. In clinical practice, in-hospital mortality with IABP or VA-ECMO often exceeds 50–60%. Given the heterogeneity of AMI-CS, rapid identification of appropriate tMCS candidates and personalized therapy are essential. Management guided by individual patient profile, hemodynamic stage and neurological status, supported by multidisciplinary shock teams, may improve timely triage, device selection and outcomes. This review emphasizes the need for individualized, protocol-driven care within structured shock systems to optimize tMCS use in AMI-CS. Full article

Background: While Immune Checkpoint Inhibitors (ICIs) have significantly improved outcomes in lung cancer (LC), clinical responses remain heterogeneous. Static tissue biomarkers, like PD-L1, and tumor mutational burden (TMB) are limited by intratumoral heterogeneity and the inability to track temporal changes. This review aims to evaluate the current state and future potential of liquid biopsy as a dynamic tool for patient selection, treatment monitoring, and the identification of resistance mechanisms in LC immunotherapy. Methods: A literature search was conducted in the PubMed database up to March 2026. We identified 65 eligible publications, including clinical trials, observational studies, and systematic reviews, focusing on liquid biopsy analytes such as circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), exosomes, and soluble immune mediators. Results: Liquid biopsy provides a “pooled” representation of the total tumor burden, overcoming the spatial limitations of tissue biopsy. Key findings include that dynamic changes in ctDNA and bTMB can predict molecular progression weeks before radiological assessment; blood-based PD-L1 monitoring (soluble, exosomal, or on CTCs) correlates with survival outcomes and offers a real-time readout of immune checkpoint activity; novel markers like tumor-macrophage fusion (TMF) cells and cytokine signatures provide unique insights into the systemic immune microenvironment. Conclusions: Liquid biopsy is evolving from a complementary diagnostic tool into a central pillar of precision immuno-oncology. Although technical standardization remains a challenge, the integration of multi-omic blood-based biomarkers represents the future of personalized lung cancer management. Full article

Acute hypoxemic respiratory failure (AHRF) represents one of the most common and clinically challenging indications for invasive mechanical ventilation in the intensive care unit, characterized by profound etiological heterogeneity that demands accurate diagnosis to guide treatment. While clinical history, physical examination, and laboratory data remain essential, they are often insufficient to reliably discriminate among conditions such as acute respiratory distress syndrome (ARDS), cardiogenic pulmonary edema, and pneumonia—particularly in mechanically ventilated patients. Lung imaging has therefore emerged as an indispensable complement to clinical assessment. In this narrative review, we systematically describe the physical principles, clinical applications, and limitations of the imaging modalities currently available in critical care: chest X-ray (CXR), computed tomography (CT), lung ultrasound (LUS), electrical impedance tomography (EIT), and positron emission tomography (PET). CXR remains the most widely used bedside tool but is constrained by low sensitivity and significant interobserver variability. CT is the gold standard for morphological and quantitative lung phenotyping, enabling the assessment of recruitability, baby lung characterization, and the identification of complications, but requires patient transport and exposes patients to ionizing radiation. LUS offers real-time, bedside evaluation of aeration with high diagnostic accuracy for pneumothorax and pleural effusion, and is increasingly integrated into revised ARDS diagnostic criteria. EIT enables continuous, radiation-free monitoring of regional ventilation distribution and positive end-expiratory pressure (PEEP)-guided titration directly at the bedside. While PET provides unparalleled quantification of regional inflammation and ventilation-perfusion mismatch, it currently remains a purely investigative research tool. Finally, we discuss emerging technological and AI-driven advances—including dual-energy CT, next-generation EIT, and deep learning algorithms—that are poised to transform lung imaging from a passive diagnostic tool into an active, personalized guide to respiratory management. Full article

WiFi sensing has gained widespread attention as a promising technology, owing to its non-intrusiveness, strong privacy-preserving characteristics, and cost-effective deployment, enabling diverse application scenarios. In addition, the stable spatial characteristics and symmetry-related patterns exhibited by human body postures in WiFi signal propagation provide new possibilities for robust person identification. In traditional WiFi-based person identification technologies, although gait recognition has achieved certain success, it is complex to operate and limited in application scenarios, increasing the constraints on recognition. This issue becomes more pronounced in large-scale user scenarios, where the system performance tends to degrade and exhibit instability. To overcome these challenges, we introduce a new person identification system called WiPID. The WiFi signals extracted from the static postures of users are treated as a “biometric fingerprint” for identity verification. An end-to-end deep learning framework is utilized by WiPID to process WiFi signals, and a convolutional autoencoder is adopted to preprocess the signals directly, effectively reducing redundant information and greatly simplifying the WiFi data processing. Furthermore, the integration of a multi-scale feature extraction module improves the system’s ability to capture discriminative features. The proposed system not only reduces operational complexity but also extends its applicability to a wider range of scenarios, thereby enhancing recognition performance. In an experiment involving 50 volunteers, WiPID achieved an average recognition accuracy of up to 98%, demonstrating the method’s suitability for large-scale person identification scenarios. In addition, a real-time identification experiment has been conducted on PCs and commercial WiFi devices. Experiments have proven that WiPID can achieve real-time person identification on Internet of Things devices, further validating its feasibility and stability in practical applications. Full article

The increasing demand for personalized conservative treatment of idiopathic scoliosis (IS) highlights the need for objective and continuous monitoring of corrective forces during brace therapy. This study aims to evaluate the feasibility and clinical relevance of a smart orthopedic brace equipped with integrated force sensors for long-term biomechanical assessment. Three female patients with different types of idiopathic scoliosis were treated using a custom-designed thoracolumbosacral orthosis incorporating four flexible pressure sensors, enabling real-time and long-term recording of corrective forces at key anatomical locations. Sensor data were analyzed in relation to brace-wearing adherence, patient activity, and radiological outcomes assessed using Cobb angle measurements. The results demonstrated substantial variability in force distribution and wearing patterns among patients, which was associated with differences in treatment effectiveness. Higher and more stable corrective forces near curve apices were generally accompanied by improved radiological outcomes, whereas irregular brace use and uneven pressure distribution limited therapeutic effects. Long-term monitoring enabled identification of insufficient correction zones and adherence issues. In conclusion, the proposed sensor-based orthotic system provides clinically relevant information on force distribution and brace use, supporting individualized therapy optimization. These findings indicate that smart braces can enhance clinical decision-making and contribute to more effective and personalized scoliosis management. Full article

Pancreatic cancer (PC) remains one of the most aggressive and lethal malignancies worldwide, largely due to late diagnosis, aggressive biology, limited therapeutic options and responsiveness. Conventional diagnostic and monitoring strategies, including imaging and serum biomarkers such as CA 19-9, provide limited sensitivity for early detection and suboptimal accuracy for the dynamic assessment of treatment response and disease evolution. These limitations highlight the urgent need for innovative, minimally invasive approaches capable of improving patient stratification and guiding personalized management. In this context, liquid biopsy has emerged as a promising, minimally invasive approach able to capture tumor-derived molecular information through the analysis of circulating cell-free nucleic acids, including circulating cell-free DNA (cfDNA) and circulating cell-free RNA (cfRNA). Released into the bloodstream by tumor cells, these analytes offer a real-time and comprehensive snapshot of tumor biology, capturing genetic, epigenetic, and transcriptional alterations through a simple blood draw. Liquid biopsy-based analyses hold significant potential for early detection, prognostic assessment, therapeutic decision-making, monitoring of minimal residual disease, and identification of resistance mechanisms. This review discusses the current state of research on circulating cell-free nucleic acids in PC, highlighting their biological basis, methodological approaches, clinical potential, and the challenges limiting their widespread implementation. By underscoring their translational relevance, we aim to outline how integrated liquid biopsy strategies, alongside the need for standardization and cross-study harmonization, may contribute to a more precise and dynamic approach to PC management. Full article

Wearable device-based Human Activity Recognition (HAR) is widely used in health management, rehabilitation, and personal safety. While contemporary HAR research effectively classifies a wide range of discrete activities, there remains a significant gap in organizing these heterogeneous motions into a structured intensity framework suitable for continuous risk assessment. Furthermore, many high-performing models rely on computationally intensive architectures that hinder real-time deployment on resource-constrained wearables. We propose an on-device method for estimating five-level activity intensity in real time using only accelerometer signals from a commercial smartwatch. To bridge the gap between simple identification and intensity modeling, 13 dynamic and emergency-like wrist motions were integrated with 11 daily activities from the PAMAP2 dataset, yielding 21 activities mapped onto an ordinal five-level intensity scale. A finetuned Multi-Layer Perceptron (MLP) classifier trained on this integrated dataset achieved 0.939 accuracy and a quadratic weighted kappa (QWK) of 0.971. The model was deployed on a Galaxy Watch 7, achieving $<1$ ms inference latency and a size $<0.1$ MB, confirming real-time feasibility. This approach demonstrates that organizing diverse activities into a lightweight, intensity-aware framework provides a robust foundation for safety-aware monitoring systems under real-world, on-device constraints. Full article

Minimal residual disease (MRD) refers to the persistence of low-level malignant cells or tumor-derived nucleic acids that remain after curative-intent therapy and are undetectable by conventional diagnostic methods. In oncology, MRD has emerged as a powerful biomarker with well-established prognostic value in hematologic malignancies and rapidly expanding relevance in solid tumors. Advances in sensitive detection technologies, including multiparameter flow cytometry, quantitative real-time polymerase chain reaction, next-generation sequencing, and digital polymerase chain reaction, have enabled the identification of residual disease at the molecular level, often preceding clinical or radiological relapse. Beyond its conventional role as a binary indicator of treatment response or cure, MRD is increasingly recognized as a dynamic longitudinal biomarker that supports personalized disease management. Within this evolving paradigm, patient-informed MRD strategies that incorporate tumor-specific molecular profiling and serial monitoring, particularly through circulating tumor DNA, offer the potential to guide treatment adaptation, including escalation, de-escalation, maintenance optimization, and surveillance strategies across both hematologic and solid malignancies. In this review, we summarize the biological basis of MRD, current and emerging detection methodologies, and clinical applications across cancer types, with a focus on patient-informed approaches. We also discuss key limitations, including assay standardization, biological variability in solid tumors, and the lack of clearly defined actionability thresholds. Finally, we highlight future directions for integrating MRD with multi-omics and AI-driven analytical frameworks to enable adaptive, risk-informed cancer management and advanced precision oncology. Full article

The risks associated with hazardous materials (HazMat) transportation exhibit typical characteristics of chain-like distribution, spatiotemporal regularity, and individual heterogeneity. A personalized trip-chain feature spectra recognition framework for HazMat vehicles is proposed to enhance the capability to assess and analyze individual risks using vehicle positioning data. The proposed framework addresses the challenges of deriving personalized risk feature maps arising from missing real-time trajectory data, complex sub-trip-chain segmentation, and the extraction of personalized risk feature representations. An improved conditional Wasserstein Generative Adversarial Network (WGAN) model is initially developed to impute trajectories with missing positional data, and it can robustly reconstruct trajectories with large-scale missing segments by integrating a multi-head self-attention mechanism and a gradient penalty. A two-layer clustering algorithm, K-Means-multiplE-THreshOlds-adaptive-DBSCAN (KMETHOD), which combines an adaptive mechanism with threshold rules, is subsequently designed to identify the dwell time and related spatial attributes of dwell points along vehicle trips. A BERT-based model is incorporated to filter Points of Interest (POIs) around dwell points, which enables the extraction of their detailed location semantics and trip characteristics and thus supports trip chain identification and segmentation. A threshold-activated multilayer trajectory feature-map method (TAFEM) is constructed to generate feature maps for each trip chain. The Liquefied Natural Gas (LNG) transportation trajectory data from Guangdong Province is selected to evaluate the effectiveness of the proposed methods. The experimental results demonstrate that the proposed framework can effectively identify trip chains and generate their corresponding feature maps. The trajectory imputation model achieved the Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Dynamic Time Warping (DTW) of 2.34–3.33, 6.05–7.74, and 0.74–1.21, respectively, across different missing-rate scenarios, outperforming other benchmark models. The identification accuracy of dwell-point duration and location reaches 98.35%. The BERT-based method achieves a maximum accuracy of 92.83% in origin–destination (OD) point recognition, effectively capturing comprehensive trip-chain information. TAFEM accurately characterizes the spatiotemporal distribution and potential causal factors of personalized HazMat transportation safety risks, providing a reliable foundation for risk identification and safety management strategies. Full article

Road safety depends heavily on the timely identification and repair of potholes; however, detecting potholes is challenging due to various lighting and weather conditions. This work presents an attention-enhanced object detection framework for aerial pothole detection design that relies on a pre-trained backbone, YOLOv8, and a custom feature-extraction network, the Feature Pyramid Network (FPN). An enhanced detection head is used to make the model aware of discriminative areas in space to get accurate localization of a pothole to overcome the major limitations of the standard YOLOv8 used in aerial road inspection, irrespective of the road surface. The underlying architecture incorporates a purpose-built data layer and a preprocessing engine that can accommodate scenarios such as seasonal changes and bad weather. To further enhance learning dynamics, a customized loss function and a new optimizer framework are incorporated to improve convergence towards overall detection reliability. Specifically, a custom differential optimizer that uses layer-wise adaptive learning rates and momentum-based gradient updates to help suppress false positives and accelerate convergence. Conversely, the IoU-based personal loss function, combined with real-time validation, stabilizes training across a range of road conditions. A major feature of the proposed system is its ability to process aerial imagery from unmanned drone platforms. Empirical analysis proves a good result: an average precision of 0.980 with the IoU of 0.5 and an F1-score of 0.97 with a confidence threshold of 0.30. Precision is high (0.97 at the 90-percent confidence level). These metrics show how well the model will be able to balance false positives and false negatives—a critical need in a safety-critical deployment. The results make the framework a potential, scalable, and reliable candidate for integrating smart transportation systems and autonomous vehicle navigation. Full article